A Comparison of Neural network and Fast Fourier Transform-based Approach for the State Analysis of Brain

Electroencephalogram (EEG) signals are used in medical field to assess the functional states of the brain. The state of brain is evaluated in the specified frequency domain by using EEG signals. Thus monitoring the state of brain involve notable characteristic. In this paper we propose a new neural network based technique to address those problems. We show that a feedforward, multi-layered neural network can conveniently capture the state of the brain in its connection weight-space, after a process of supervised training. The performance of the proposed method is investigated with the EEG signals associated with the state change of the brain, and also compared with a fast fourier transform (FFT)-based spectral estimation technique in spectral representation of a signal against time.